Page 8 of 15
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 06:48
by kludge
Maximilian wrote: 21 Mar 2019 23:20
Could this be the problem?
In the a600 the BR is tied to VPA and those are tied to a 1k resistor.
So if you pull down BR you are also pulling down VPA......not a problem during startup but maybe during normal running?
Also the 1k pullups are a lot stronger then the 470/4k7 ohm resistors on the A500.
Knipsel.PNG
WTH. I'm pulling BR low with a 220 ohm resistor, so it should be stronger than the pullup. But I had completely missed the VPA connection. What does VPA do?
Edit: It doesn't sound like a great idea to have VPA permanently asserted. So, right now the DIP64 VPA is being pulled low by the 220 ohm resistor on the PLCC BR line... That probably needs to change. Next: Break the connection between the DIP64 VPA and PLCC VPA and connect the DIP64 VPA to the DIP64 BR. Or am I being stupid here?
Screenshot 2019-03-22 at 07.54.31.png
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 07:06
by kludge
If I had a time machine I would go back in time and tell Motorola to add a chip enable to the 68k line of CPUs.
...but Commodore would probably have connected it in some awkward way preventing me to use it anyway.
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 08:26
by Maximilian
There are turbo boards for the a600, but they probably use a small processor for this.
So I would not go and change your motherboard, otherwise you could just remove the plcc 68k and be done with it.
Isn't the VPA used for the Cia's etc?, didn't have time to check that.
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 08:46
by kludge
It looks like VPA is just pulled up together with BR, so I don’t think it used by anything else. But I’ll check again after work.
No, I want to handle everything on the adapter, ideally. No hardware modifications on the motherboard or the accelerator.
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 09:14
by terriblefire
VPA/VMA/E are how the 68000 talks to the CIAs.
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 09:20
by kludge
terriblefire wrote: 22 Mar 2019 09:14
VPA/VMA/E are how the 68000 talks to the CIAs.
It looks like it's only being pulled high.
Screenshot 2019-03-22 at 10.19.17.png
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 09:33
by terriblefire
VPA, VMA and E are the 6800 (not 68000) control lines. You shouldn’t touch any of these lines. The CIA use these to terminate bus cycles.
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 09:40
by kludge
terriblefire wrote: 22 Mar 2019 09:33
VPA, VMA and E are the 6800 (not 68000) control lines. You shouldn’t touch any of these lines. The CIA use these to terminate bus cycles.
But if I'm asserting BR I'm also touching VPA as they are tied together on the A600 (and pulled high, without any connections elsewhere if PCB explorer is to be trusted), so I don't understand how I'm supposed to not mess with it. My thinking is that by pulling VPA high on the DIP64 section of the adapter I'm just replicating what is already done on the A600 motherboard while still being able to assert BR.
Edit:
Screenshot 2019-03-22 at 10.43.00.png
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 10:24
by terriblefire
If you mess with these lines and/or don’t connect them to the Amiga then the machine won’t work.
Why would you pull them up if the Amiga already does this anyways??
EDIT: these are real signals between the Amiga chipset and the cpu you’re adding. Don’t mess with them.
Re: Compatibility issues with TF530/TF534 and A600?
Posted: 22 Mar 2019 10:29
by kludge
terriblefire wrote: 22 Mar 2019 10:24
If you mess with these lines and/or don’t connect them to the Amiga then the machine won’t work.
Why would you pull them up if the Amiga already does this anyways??
Because I want to pull BR low for the bus arbitration, and by doing that VPA would also go low as they are tied together on the A600. I can’t see how these could be used to communicate with the CIAs as there doesn’t seem to be any traces between the chips. At all.
This is weird.